Three-Dimensional Echocardiography: Advancements in Qualitative and Quantitative Analyses of Mitral Valve Morphology in Mitral Valve Prolapse.

Degenerative mitral valve disease (MVD) is the leading cause of organic mitral regurgitation (MR), one of the most common valvular heart disease in western countries. Substantial progresses in the surgical treatment of degenerative MVD have improved life expectancy of patients with significant MR. However, prognosis, surgical decision and timing of surgery strongly depend on the accurate characterization of mitral valve (MV) anatomy and pathology and on the precise quantification of MR. Three-dimensional (3D) echocardiography, a major technological breakthrough in the field of cardiovascular imaging, provides several advantages over two-dimensional (2D) imaging in the qualitative and quantitative evaluations of MV apparatus. In this review, we focus on the contribution of this new modality to the diagnosis of degenerative MVD, the quantitative assessment of MR severity, the selection and monitoring of surgical and percutaneous procedures, the evaluation of procedural outcomes. The results of a systematic and exhaustive search of the existing literature, restricted to real-time 3D echocardiography in adults, are here reported.

INTRODUCTION

Moderate or severe mitral regurgitation (MR) is the most frequent valve disease in the USA and the second most common form of valvular heart disease needing surgery in Europe.[12] Degenerative mitral valve disease (MVD) is the leading cause of organic MR in western countries. Its mechanism is primarily characterized by mitral valve prolapse (MVP) due to excessive leaflet movement (≥2 mm abnormal systolic movement beyond the saddle-shaped annular level), as defined by Carpentier (type II).[3] The anatomic lesions associated with the degenerative MR etiology encompass a wide pathological spectrum, from fibroelastic deficiency (FED) to extensive myxomatous disease. Carpentier described FED as a condition associated with a fibrillin deficiency that often leads to the rupture of one or more thinned and elongated chordae. This process usually, but not exclusively, involves the middle scallop of the posterior leaflet with the myxomatous degeneration limited to the prolapsing segment.

In contrast to FED, myxomatous disease, often called Barlow disease (BD), is characterized by diffuse excess tissue, with large valves comprising multiple degenerated segments presenting as thick, mostly with elongated chordae; severe mitral annulus (MA) dilatation associated with different degrees of annular calcification is a frequent finding in patients with BD, and subvalvular fibrosis and calcification of the papillary muscles, usually the anterior, may also occur. Myxomatous degeneration is defined by the accumulation of mucopolysaccharides responsible for the thickening and ‘proliferative’ aspect of valvular tissue. Even if molecular disorders and pathophysiologic pathways may not be uniform, there is a continuum between the different anatomic aspects of degenerative MR.

Surgery is the treatment of choice in degenerative MR, improving symptoms and preventing heart failure. Since the 1970s, mitral valve (MV) repair has emerged as the optimal treatment, preferred to traditional valve replacement, as it improves outcome and reduces mortality of patients with severe organic MR by about 70%. Recent studies and guidelines have underlined the importance of early surgical intervention to preserve long-term left ventricular (LV) function in severe MR.[45] Indeed, the best short- and long-term results are obtained in asymptomatic patients operated in advanced repair centers with low operative mortality (<1%) and high repair rates (≥80-90%). These results emphasize the importance of early detection and detailed assessment of valve morphology and MR.

Moreover, in the recent years there has been an increasing interest towards non-surgical solutions to severe MR, with the percutaneous MitraClip system, mimicking the surgical Alfieri stitch by placing a clip on the beating heart, being the most well established, with results not inferior to surgical correction in selected patients.[678] Patient selection is a crucial step of the MitraClip procedure and several anatomic criteria should be fulfilled to ensure the suitability of the device to MV anatomy.

In this complex scenario, echocardiography is the method of choice to evaluate patients with degenerative MR; it allows the accurate assessment of MR severity, etiology, mechanisms and anatomic lesions, LV and left atrial remodeling, and consequently defines the indication and the probability of successful MV repair.

Three-dimensional (3D) echocardiographic imaging has represented a major innovation in cardiovascular ultrasound. Recent guidelines stated that real-time three-dimensional (RT3D) imaging modalities are useful in the presentation of realistic views of heart valves and are ideal for interrogating the anatomy and function of each of the individual components of the MV apparatus, including annulus, leaflets, chordae and papillary muscles.[910] Both qualitative and quantitative evaluations of degenerative MVD have been substantially improved by 3D echocardiography.[111213]

METHODS

All publications considered for this review were identified accessing the Medline database via Pubmed. The search strategy employed a number of free-text keywords and controlled vocabulary terms including, but not limited to, the following concepts: 3D echocardiography; mitral valve; degenerative mitral valve disease; mitral valve repair; percutaneous mitral valve repair. Fields of search were limited to the Title and Abstract and restricted to the English language. All the retrieved studies were screened for inclusion using a hierarchical approach assessing title, abstract and manuscript. Studies using 3D multiplanar reconstructive echocardiography or enrolling pediatric patients only were excluded. Additional searches were conducted after screening the references of all selected studies.

3D qualitative and quantitative evaluations of degenerative mitral valve disease

As a direct consequence of a series of technological breakthrough, such as the introduction of the 3D matrix technology with its improved spatial resolution compared to previously available sparse array transducers, RT3D echocardiography demonstrated feasibility and accuracy in MV assessment. The enhanced visualization of morphologic and pathologic aspects of the MV apparatus in degenerative MR has contributed to our knowledge of MV pathology, with unquestionable benefits compared to traditional two-dimensional (2D) echocardiography.[111213]

Qualitative morphological evaluation of the MVP may be accurately assessed both by 3D transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) and the recent data showed that RT3D, either TTE or TEE, is superior to the corresponding 2D techniques in the description of MV pathology [Figure 1].

Figure 1

Example of head-to-head comparison between 3D TTE (left) and 3D TEE (right) datasets in two cases of A2 prolapse (above) and P2 prolapse (below), imaged from the surgical view (left atrium)

3D evaluation may be easily integrated into a standard 2D examination in the assessment of qualitative morphology of the MV.[14] Indeed, the possibility of visualizing MV leaflets, commissures, annulus calcifications and subvalvular structures in different and unique planes, both from the atrium or ventricle, with access to “en face” views, enhances the understanding of this complex apparatus.

Diagnostic accuracy of RT3D TTE vs surgical inspection in the identification of prolapsing segments has been investigated by Tamborini et al., in a series of consecutive 200 cases. The overall accuracy of the method was high (sensitivity 92.5%, specificity 96%, accuracy 95%); sensitivity was slightly lower for antero-lateral commissure, P1 and A1, while specificity was very high in all the segments. Even focusing on the differences in RT3D TTE accuracy in the setting of simple (isolated P2 prolapse or P2 associated with P1 or P3) and complex cases (lesions involving the entire posterior leaflet, lesions of the anterior leaflet, bileaflet prolapse, and commissural involvement), the accuracy was very high, 98% and 93% respectively, even if slightly lower in complex lesions.[15] All these data were in accordance with previous findings.[16]

As it concerns RT3D TEE, there is a substantial body of literature demonstrating the additional value of 3D over 2D in the evaluation of patients with MVP.[171819] The majority of these studies showed that 3D echocardiographic findings correlate closely to surgical findings, achieving an exact anatomic description in approximately 90% to 95% of segments, whereas 2D echocardiography is less accurate. Grewal et al., underlined that in agreement with previous works, the incremental value of 3D modality over conventional 2D can be better appreciated in complex disease involving both leaflets or the anterior leaflet alone.[20] La Canna et al., demonstrated that RT3D TEE provided more accurate mapping of MV prolapse than 2D TEE and RT3D TTE imaging, adding quantitative recognition of dominant and secondary lesions and MV anatomy details in 222 patients undergoing surgical repair for MR secondary to prolapse.[21] Recently, Hien et al., investigated whether the known benefit of RT3D TTE over 2D TEE imaging for the evaluation of MVP would be valuable among novice echocardiographers. They demonstrated that both expert and novice echocardiographers were able to provide more accurate descriptions of the MV prolapse with RT3D than with 2D TEE images, and this benefit was three times more important for inexperienced operators.[22]

With the advances of RT3D echocardiography together with the development of dedicated RT3D software capable of providing several parameters (annular diameters and area, annular height and planarity, saddle-shaped morphology, leaflet size, coaptation geometry), high-resolution, full-volume imaging and quantification of morphology of the entire mitral apparatus have become feasible [Figure 2]. Several studies demonstrated that the use of RT3D allows improving the evaluation of the MV apparatus both qualitatively, as previously illustrated, but also quantitatively. In fact, the MV has a complex 3D morphology and attempts to use 2D images to comprehend a complex 3D structure have provided an incomplete picture. In particular, the normal MA is characterized by a nonplanar saddle shape that has been suggested to be important for alleviating the mechanical stress on mitral leaflets and chordae tendinae imposed by LV pressure.[2324] Moreover, the geometry of leaflet surface and coaptation, magnitude of leaflet billowing, and deformation of subvalvular structures are important morphological features in MVP and they can be only partially assessed with 2D imaging. Chandra et al., hypothesized that RT3DE-derived measurements of valvular anatomy could be used to characterize MVD in an objective way. Morphological analysis in the MVD group revealed a progressive increase in multiple parameters from normal subjects to FED to BD, allowing an accurate diagnosis of these entities. The strongest predictors of the presence of MVD included billowing height and volume. Three-dimensional billowing height with a cutoff value of 1.0 mm differentiated MVD from normal subjects, and billowing volume with a cutoff value of 1.15 mL differentiated between FED and BD. These findings could support the adoption of preoperative automated clinical decision-making strategies based on reliable quantitative parameters, with important surgical implications.[25]

Figure 2

Example of 3D TEE and corresponding reconstruction of mitral annulus and leaflets in a patient with fibroelastic deficiency (left) and in two patients with Barlow disease (right). In the first case, reconstruction clearly shows (red surface) the P2 prolapsed segment. The two Barlow cases with the same method are characterized by multiple scallop involvement of both leaflets identified by the red surfaces

Moreover, Grewal et al., investigated mitral annular size, shape, and motion 6 times over the cardiac cycle using RT3DE in patients with degenerative MVD compared with normal subjects and patients with ischemic MR and demonstrated that degenerative MVD annulus is still a dynamic structure but behaves considerably different from normal patients with loss of early-systolic area contraction and saddle-shape deepening. Subsequent area enlargement may contribute to mitral incompetence.[26]

Maffessanti et al., characterized the MV 3D morphology using RT3D TEE in the presence of severe MR associated with FED or BD, immediately before and after MV repair, in comparison with a control group. They found that the annular enlargement after MR is not isotropic, being more pronounced in the anteroposterior and intercommissural directions rather than along the mitral height, resulting in flattening of the MV in both FED and BD groups with higher values in the BD group. Postoperatively, annular diameters and MV area were significantly smaller compared with preoperative values; similar changes were found in the exposed areas of both leaflets.[27]

More recently, Lee et al., demonstrated that patients with MVP-related MR had more dilated mitral annulus, a flattening of annular saddle shape, redundant leaflet surfaces, greater leaflet billow volume and billow height, longer lengths from papillary muscles to coaptation and more frequent chordal rupture compared to normal subjects and non-prolapse MR patients. Interestingly, the association between annular flattening and the increased frequency of chordal rupture suggests that the loss of the annular saddle shape may predispose to chordal elongation and rupture as a result of excessive chordal tension.[28]

QUANTIFICATION OF MR

The prognosis and the best timing for surgical treatment in degenerative MVD strongly depend on the accurate quantification of MR severity, which claims for accurate, feasible and reproducible techniques.[2930] Despite these premises, the quantitative assessment of MR remains an open issue in clinical practice. Recent technological development has made RT3D echocardiography the ideal technique for the evaluation of MR severity, in particular, in all those cases where the hypotheses at the basis of 2D estimate are not met. MR grading can be obtained, in accordance with standard 2D echocardiography, by means of vena contracta area (VCA), effective regurgitant orifice area (EROA) using 3D proximal isovelocity surface area (PISA) or MR volume measurement [Figure 3]. Detailed and comprehensive discussions on MR quantitative evaluation have been recently reported by Thavendiranathan et al.[3132]

Figure 3

Two examples of the complementary use of 3D color Doppler in two different patients with a more common and typical patterns (left case P2 prolapse) and an uncommon case with multiple scallop involvement. 3D TTE color Doppler dataset (left) clearly facilitates the identification of the origin of the jet (P2) and its measurement while in the complex case (right) TEE identifies multiple regurgitant jets

3D VCA can be easily obtained, without any a priori shape assumption, acquiring a 3D color Doppler dataset, and then cutting the volumetric dataset till the best en face view of the regurgitant jet is obtained; VCA is measured as the MR jet planimetry. Several studies have shown VCA estimated by RT3D echocardiography to be more accurate and reproducible than 2D echocardiography-derived VCA, when compared with other techniques such as cardiovascular magnetic resonance.[333435] Moreover, 3D-VCA measurements appear accurate even in asymmetric regurgitant orifices and unaffected by the etiology or eccentricity of MR. A cutoff of 0.41 cm2 using 3D-VCA showed 82% sensitivity and 97% specificity in differentiating moderate from severe MR.[36] Further, using the planimetric 3D-VCA and sampling the regurgitant jet with the continuous wave Doppler, MR volume can be obtained multiplying the anatomical regurgitant orifice area and the velocity time integral, similarly to what is done routinely using 2D echocardiography.

The major limitation affecting the 3D-VCA technique is related to the limited temporal and spatial resolutions, which may prevent the accurate measurement of VCA.[3637]

The anatomic severity of MR can be assessed by 3D color Doppler using the PISA technique. The theoretical benefit of the 3D PISA technique is the ability to measure the 3D surface of the proximal flow convergence region (PFCR) or to obtain the largest radius of the PFCR using 3D navigation, possibly increasing the accuracy of the EROA calculation.[3839] Although the use of 3D data may improve the accuracy of the EROA calculation, it still requires assumptions about the shape of the PFCR.[384041] Two studies attempted to obtain a 3D surface area, one using measurement of multiple radial planes of the PFCR in an in vitro model to reconstruct subsequently the total surface area,[42] and the second using multiple linear measurements to reconstruct the 3D surface area.[43] Both methods are laborious and not practical in a busy clinical setting. Recently, a completely 3D-based method for MR quantification has been proposed and validated against cardiovascular magnetic resonance. The MR volume is obtained subtracting the left ventricular (LV) outflow from the MV inflow; both flows are calculated based on 3D color-Doppler acquisitions, using advanced algorithm for the estimate of the flow rate in planes close to the MV or LV outflow tract.[44] This can be more accurate and reproducible than 2D pulsed-wave Doppler-based methods.[45]

Despite several studies have clearly shown the benefit of RT3D echocardiography, both TEE and TTE, for the MR quantification when compared to standard 2D echocardiographic methods, the lack of extensive, independent and ideally multicentric validation studies, limits the widespread use of the above mentioned techniques.

IMPLICATIONS FOR MV SURGERY

Guidelines on management of valvular heart diseases underlined that MV repair should be considered even in asymptomatic patients when the likelihood of a successful repair is high and in presence of a low operative risk.[45] It is clear that the likelihood of a durable and safe MV repair strongly depends on the anatomy and morphology of the MV, and similarly the complexity of the surgical procedure is related to the complexity of the MV lesions.[164647] In a paper by Tamborini et al., the relationship between the extent of MV lesions and the complexity of the surgical procedures has been investigated. The results clearly demonstrated that simple lesions, as assessed by RT3D echo, are very likely to be treated with simple techniques; while in almost half of the patients with complex lesion, complex surgical repair was performed by the surgeon.[15] Moreover, Biaggi et al., showed that quantitative MV characteristics, as assessed by RT3D TEE rather than 2D TEE, determined the complexity of MV repair, being involvement of the anterior leaflet, bileaflet prolapse, severe annular calcification, increased annular dimensions and leaflet height anatomic predictors of a lower likelihood of successful repair.[48] Thus, an accurate pre-operative RT3D TTE evaluation, eventually using ad hoc post-processing software, may facilitate surgical planning allowing a tailored approach to each case.

Several publications have underlined the importance of MV annuloplasty during repair of degenerative MVD to optimize repair durability.[495051] Debate persists regarding the superiority of partial band versus complete ring devices. Suri et al., using RT3D TEE, demonstrated the presence of an enlarged posterior MA in patients with significant MR due to degenerative MVD, while there was no evidence that the intertrigonal distance is abnormal in these patients leading to the conclusion that posterior annular reduction with a flexible device at the time of MV repair is important, and that altering the anterior intertrigonal portion of the MA is unnecessary.[52] Caiani et al., investigated, by RT3D TTE, the changes in MA dynamics and the long-term effects induced by annuloplasty in MV organic prolapse undergoing repair using either an incomplete flexible band or a complete semi-rigid ring. Both rings showed similar dynamic changes and, at 6 months, the MA area change during the cardiac cycle was depressed compared to control groups, independently of the implanted prosthetic ring. They concluded that the main factor affecting MA function is the undersizing due to the ring, which restricts MA geometry and limits the natural MA motion.[53]

Most of the investigation focused on the MV and based RT3D echocardiography aimed at evaluating MV morphology as an isolated structure. However, Veronesi et al., investigated, using custom software, the relationship between MV and the aortic annulus. They demonstrated that in patients with MR, the dynamic mitral-aortic coupling was preserved despite the altered morphology and function of the MV. In these patients, MV repair with annuloplasty led to smaller and less pulsatile MA, with altered spatial displacement of the MV-aortic annulus complex.[54]

IMPLICATIONS FOR MV PERCUTANEOUS REPAIR

Despite surgical correction of MR has demonstrated optimal outcomes in terms of efficiency, safety and residual MR, it is often not feasible in high-risk patients.[2355] According to the results of the EuroHeart survey, approximately half of the symptomatic patients with severe MR of functional or degenerative origin are denied surgery, and the likelihood of surgery denial increases with LV dysfunction, age and presence of comorbidities. Therefore, several interventional options have been investigated during the last decades. Chiam et al., classified the proposed technologies into those targeting the chordae, the leaflets or the MA, both indirectly via the coronary sinus or directly true an edge-to-edge repair.[56] Among the latter, the MitraClip system is the more diffused and studied, and is currently under CE mark in Europe and investigational use in the USA and parts of Asia. The MitraClip system mimics the surgical MV repair procedure introduced by Alfieri, who treated a patient with an anterior prolapse by placing a pledgeted stitch to approximate the middle portions of the MV leaflets, creating a double orifice MV and successfully reducing the MR.[57]

The EVEREST trials (I and II) and the pivotal studies established the feasibility, safety and hemodynamic improvements in the large majority of patients, despite less effective at reducing MR than conventional surgery and thus with a higher rate of reintervention and/or surgical operation.[67858596061]

Currently, catheter-based MV clip repair is guided by standard, 2D TEE.[62] However, 2D TEE is limited in the direct visualization of major MV pathology, the assessment of the 3D location of the catheter-clip system during the procedure, and the effect of the procedure on 3D MV morphology.

RT3D TEE was found helpful in several catheter-based procedure such as closure of atrial septal defects and transcatheter aortic valve implantation and, providing a unique “en face” visualization of MV anatomy and MR orifice, that is demonstrated to facilitate the transcatheter closure of mitral prosthesis paravalvular leaks.[1963646566] Biner et al., hypothesized that the combined use of 2D and 3D TEE would also have additional value over 2D imaging in evaluating MR for catheter-based MV clip repair and guidance of the procedure. They demonstrated that combined 2D and 3D TEE imaging was associated with a shorter time to first clip deployment and with a general reduction in the total procedure time when compared to traditional 2D TEE guidance. They also stressed that RT3D TEE is valuable in determining the precise orientation of the clip arms guiding the interventional cardiologist till the clip is positioned perpendicularly between the two central scallops, without the need for multiple verifications and cross-checks of 2D planes.[67] Moreover, the use of 3D color Doppler could allow a better identification of the site of origin and the degree of any residual MR. These observations are crucial when a second clip has to be implanted. In this regard, Armstrong et al., identified thickened anterior mitral leaflet and more severe MR to be the echocardiographic predictors associated with a higher likelihood of a dual clips procedure.[68]

Recently, Faletra et al., showed side-by-side images obtained by fluoroscopy, 2D TEE, and RT 3D TEE. While essential data during the procedure were provided by the use of the 2D standard imaging techniques and the capture of mitral leaflets was always guided by 2D TEE, they demonstrated that in almost every step of the MitralClip procedure, RT 3D TEE may provide new additional useful data: the more precise anatomic information, the fine details of the devices and the precise relationship of catheters/clip with surrounding anatomic structures, may enhance the confidence of imaging interpretation and eventually improve the efficiency of the procedure.[69]

A crucial step during the percutaneous edge-to-edge repair of the MV is the assessment of clip attachment to the MV leaflets, which may be challenging using the 2D TEE. In the EVEREST I trial, single leaflet clip detachment occurred in 9% of the patients, demonstrating the need for new techniques that allow a more reliable assessment of clip attachment.[60] Braun et al., developed a novel method for the assessment of clip attachment using intraprocedural 3D TEE. They tried to quantify exactly the portion of MV leaflets fixed into the clip by measuring the distance between the lowermost part of the leaflet and the top edge of the clip. They found that the frequency of clip complications (partial clip detachment or displacement) was higher in patients imaged by 2D TEE compared to patients imaged by 3D TEE.[70] Hence, 3D imaging using volumetric analysis in addition to standard techniques may help to confirm a secure clip attachment during the MV clipping procedure.

RT3D TEE data of 55 patients (14 patients with degenerative MVD) acquired during the MitraClip procedure immediately before and after clip placement have been recently published with the aim of assessing changes in annulus diameter and area. The results demonstrated that MitraClip can produce immediate reductions in MA size and tenting in functional MR. In contrast, similar changes were not observed in organic MR, and this different remodeling between the two etiologies of MR may be important to improve procedural strategies.[71]

Scandura et al., demonstrated that the MitraClip system led to left cardiac chamber reverse remodeling, with significant improvement in LV size and function, similarly to what observed in surgical restoration, both in patients with prolapsed-related MR and functional MR.[72]

All these results support the importance of 3D imaging for pre-procedural patient selection and prediction of procedural effectiveness.

The major advantages of 3D versus 2D echo are summarized in Table 1.

Table 1

Advantages of 3D over 2D echocardiography subdivided by areas of application

Areas of application	Advantages of 3D echo over 2D	Reference
Diagnosis	Precise anatomic localization of prolapsing mitral valve segments	Pepi et al., (2006)[17], Sharma et al., (2007)[16], Lang et al., (2011)[12]
	Accurate segmental analysis of mitral valve prolapse in complex lesions	Grewal et al., (2009)[20], Tamborini et al. (2010)[15], La Canna et al. (2011)[21]
	Easy interpretation of 3D images with their true-life en face display of the mitral valve	Hien et al., (2013)[22]
	Quantitative analysis of mitral apparatus geometry	Chandra et al., (2011)[25], Grewal et al., (2010)[26], Maffessanti et al., (2011)[27], Lee et al., (2013)[28]
	Enhanced quantification of MR	Thavendiranathan et al., (2012)[31], Thavendiranathan et al., (2012)[32]
Surgical Implications	Preoperative assessment of the feasibility and durability of MV repair	Tamborini et al., (2010)[15], Biaggi et al., (2012)[48]
	Improved planning of surgical strategy	Suri et al., (2009)[52]
	Definition of changes in mitral annulus morphology and dynamic after annuloplasty	Caiani et al., (2011)[53], Veronesi et al., (2012)[54]
Percutaneous Repair Implications	Refined patient selection and identification of echocardiographic predictors associated with complex procedures	Armstrong et al., (2013)[68]
	Advanced procedural guidance with the precise spatial definition of the clip arms orientation	Biner et al., (2011)[67], Faletra et al., (2012)[69]
	More reliable assessment of clip attachment to the mitral leaflets	Braun et al., (2013)[70]
	Evaluation of immediate procedural results on mitral annulus morphology	Schmidt et al., (2013)[71]
	Measurement of left cardiac chamber reverse remodeling after repair	Scandura et al., (2012)[72]

CONCLUSIONS

Three-dimensional echocardiography represents a major technological breakthrough in the field of cardiovascular imaging, as it allows visualizing cardiac structures in their morphological complexity without the need for multiple acquisitions and mental reconstruction. During the past ten years a substantial body of literature has proven the potentials of 3D TEE and TEE in the evaluation of MV prolapse. Moreover, this imaging modality has not been limited to investigational studies, but several evidences have enlightened the benefits that 3D echocardiography can add to the daily clinical practice. Furthermore, 3D echocardiography plays a pivotal role in understanding the pathophysiology of MV disease, better defining surgical strategy, guiding percutaneous procedure and improving the communication inside the heart team.